RASSEGNA ECONOMICA

2011
NR.20/2011
F ondata
nel
1 9 3 1
dal
RASSEGNA ECONOMICA
RIVISTA INTERNAZIONALE DI ECONOMIA E TERRITORIO
RASSEGNA ECONOMICA - QUADERNI
Q U A D E R N I
ISSN 0390-010X
20
Premio Rassegna Economica 2011
La Rassegna Economica – di proprietà Intesa Sanpaolo – è nata nel 1931 per volontà del Banco di Napoli. Dal
2004 è curata da SRM - Studi e Ricerche per il Mezzogiorno.
SRM è un Centro Studi collegato al Gruppo Intesa Sanpaolo ed ha come obiettivo la creazione di valore aggiunto
nel tessuto sociale ed economico del Mezzogiorno, nella sua dimensione europea e mediterranea, facendo leva sul
miglioramento della conoscenza del territorio e sulla sua capacità di proposta, anche operando in rete con altre
istituzioni di ricerca. SRM, che vanta un consolidato know-how in campo di studi e ricerche sull’economia
meridionale, costituisce un osservatorio privilegiato di valutazione e di analisi dei fattori critici e di successo delle
politiche di sviluppo del territorio. Gli studi e le ricerche sono orientati a studiare le Infrastrutture, la Finanza
Pubblica e le Public Utilities, l’Economia delle Imprese e il Terzo Settore, le dinamiche dell’Economia dei paesi
del Mediterraneo e loro interazioni con l’Italia e il Mezzogiorno. Oltre alla Rassegna Economica, SRM pubblica la
rivista Dossier Unione Europea e un focus trimestrale on-line sull’economia delle regioni meridionali.
Presidente:
Paolo Scudieri
Consiglio Direttivo:
Giuseppe Castagna, Francesco Saverio Coppola, Gregorio De Felice,
Adriano Giannola, Marco Morganti, Antonio Nucci
Collegio dei Revisori:
Danilo Intreccialagli (Presidente), Giovanni Maria Dal Negro, Lucio Palopoli
Direttore Generale:
Massimo Deandreis
Comitato di Redazione Rassegna Economica:
Pietro Alessandrini, Gregorio De Felice, Ennio Forte, Adriano Giannola,
Vincenzo Maggioni, Massimo Marrelli, Maria Teresa Salvemini Ristuccia
Soci fondatori e ordinari
SRM - Studi e Ricerche per il Mezzogiorno
via Cervantes, 64
80133 Napoli - Italy
Tel. +39 081 4935292
[email protected]
www.sr-m.it
RIFERIMENTI REDAZIONALI
Rassegna Economica
SRM-Studi e Ricerche per il Mezzogiorno
Via Cervantes, 64
80133 Napoli
Italy
Direttore Responsabile:
Massimo Deandreis
Redazione:
Salvio Capasso (Coordinatore Editoriale)
Raffaela Quaglietta (Segreteria di redazione)
Si ringrazia Marina Ripoli per la collaborazione nella fase organizzativa del Premio
Rassegna Economica.
Recapiti:
Tel: +39 081 4935292 fax: +39 081 4935289
Email: [email protected]; [email protected]
NOTE
Rassegna Economica è disponibile anche in
http://www.sr-m.it/numeri-rassegna/
versione elettronica alla pagina:
Gli iscritti alle liste di distribuzione della rivista cartacea sono tenuti ad informare
tempestivamente la redazione di eventuali cambi di indirizzo.
Gli articoli proposti per la pubblicazione devono conformarsi alle norme redazionali
disponibili al seguente indirizzo:
http://www.sr-m.it/images/stories/allegati/norme%20di%20redazione%20rassegna.pdf
Le opinioni espresse negli scritti pubblicati in questo periodico impegnano esclusivamente la
responsabilità degli autori. La riproduzione degli articoli è consentita con citazione della fonte.
Rassegna Economica
Quaderni di ricerca
Pubblicazione periodica
Associazione Studi e Ricerche
per il Mezzogiorno
Direttore Responsabile
Massimo Deandreis
N. 20
DICEMBRE 2011
SOMMARIO
Massimo Deandreis
Introduzione
5
Francesco Saverio Coppola
La Rassegna Economica 1931-2011. I suoi
primi ottant’anni
7
Paolo Veneri,
Andrea Cirilli Struttura economica, identità territoriale e
crescita nei sistemi locali del Mezzogiorno 13
Anna Di Bartolomeo Trend demografici, sviluppo economico e
dualismo Nord-Sud
33
Cosimo Magazzino,
Marco Mele
Health Care Expenditure and Economic
Growth in Southern Italian Regions: an
Empirical Analysis
51
Cristian Barra
Efficienza bancaria, fattori ambientali e
sviluppo economico: Alcune stime su micro
dati
67
Oriana Romano
Il ciclo integrato dei rifiuti: il divario Nord-Sud
e strategie di intervento
Notizie sugli autori
101
127
3
HEALTH CARE EXPENDITURE AND ECONOMIC GROWTH
IN SOUTHERN ITALIAN REGIONS: AN EMPIRICAL ANALYSIS
Abstract: Questo saggio investiga il nesso tra la spesa sanitaria e il PIL per le regioni del
Mezzogiorno d’Italia nel periodo 1980-2009 utilizzando un approccio di serie storiche. Lo studio
parte da un’introduzione e una rassegna della letteratura economica sul tema, prima di discutere i dati
utilizzati e di introdurre alcune tecniche econometriche. I test di stazionarietà e radici unitarie
rivelano che le serie della spesa sanitaria e del PIL sono entrambe I(1), per tutte le regioni. Inoltre,
troviamo ovunque una relazione di cointegrazione tra le variabili. La dinamica di breve periodo
mostra che il flusso di causalità è di tipo bidirezionale, in sei casi su otto, e anche nel lungo periodo
sussiste una relazione causale bidirezionale (o “effetto di feedback”) tra le due serie. Di conseguenza,
concludiamo che la spesa sanitaria sia un fattore limitante per la crescita economica delle regioni
dell’Italia Meridionale.
Keywords: Health policies; health care expenditure; GDP; stationarity; cointegration; causality;
South-Italy
JEL: B22; C22; I18
1. INTRODUCTION
The causal relationship between health care expenditure and economic growth has
been a well-studied topic. Health is one of essential factors for any country’s
economic development and therefore plays an important role in economy activities.
Over the past three decades, a lot of studies – using the concepts of
cointegration and Granger causality – focused on several countries and time periods.
Since the pioneering studies by Newhouse (1977) and Parkin et al. (1987), empirical
findings are mixed and, for some countries, controversial (Devlin and Hansen,
2001). The results differ even on the direction of causality and the short-term versus
long-term effects on health policies. Depending upon what kind of causal relationship
exists, its policy implications may be significant. Moreover, multiple causality studies
have been done for many countries in the world; however, few studies have been
devoted to the analysis of this nexus for the Italian case (Piperno and Di Orio, 1990;
Devlin and Hansen, 2001; Giannoni and Hitiris, 2002; Erdil and Yetkiner, 2009;
Magazzino, 2011). So, this paper examines the nexus between GDP and health care
expenditure in South-Italian regions for the period 1980-2009, using time series
methodologies on correlation, stationarity, cointegration and causality. The results
might help to define and implement the appropriate health development policies in
these regions. The data used are obtained by ISTAT.
The outline of this paper is as follows. Section 2 provides a survey of the economic
literature on the nexus between health care expenditure and GDP. Section 3 contains
an overview of the applied empirical methodology and a brief discussion of the data
51
COSIMO MAGAZZINO, MARCO MELE used. Section 4 discusses our empirical results. Section 5 presents some concluding
remarks and, finally, Section 6 gives suggestions for future researches.
2. THE DETERMINANTS OF HEALTH EXPENDITURE AND GDP
The impact of health expenditure on public finance is becoming a usual topic of
recent comments and analysis. Textbook economic theory suggests that demand for a
good/service by a utility-maximizing consumer depends on two factors: income and
relative price. Most of the studies report an income elasticity exceeding unit, implying
that health care is a luxury good. (In contrast, Wang (2009) found a cross-section
income elasticity of health care around 0.7, implying that health care is a necessity
rather than a luxury good at the state level). A point of debate among economists is
whether the public sector should intervene or not in the short-term fluctuations in
economic activity. If classical economists have always opposed such a kind of public
action, the Keynesian school of thought invoked fiscal policies to support the economy
during recessions. In fact, the classical economists believed that market forces were
able to quickly bring economies to a long-run equilibrium, through adjustments in the
labor market. Instead, the Keynesians took the fallibility of such self-regulatory
mechanisms, precisely because of rigidities in the labor market. To this end, the school
has prescribed Keynesian expansionary fiscal policies in order to avoid long slumps.
The first model on the determinants of public expenditure is “Wagner’s Law”
(Wagner, 1883). According this theory, public expenditure is essentially explained with
the evolution of GDP. As confirmed by CEIS analyses (CEIS, 2008), GDP remains the
most “important” determinant of health expenditure, as a proxy of earned economic
conditions. Newhouse (1977) found a positive effect of income on health expenditures
at the national level, assuming no price effect. Gerdtham and Jönsson (1991a, 1991b,
2000) found a strong negative effect of relative prices on quantity demanded, with a
price elasticity of -0.84; and Milne and Molana (1991) reached about a similar
empirical results. So, as has been shown in Blomqvist and Carter (1997), Di Matteo and
Di Matteo (1998), Freeman (2003), omitting relative price in regressors’ set when its
effect is significant clearly will carry out to biased estimations.
Spinks and Hollingsworth (2007) has analyzed a number of theoretical questions in
order to make some international comparisons of the technical efficiency of health
production, enlarging their study to all social policy, and not just health policies.
Wang (2009) has examined the determinants of health expenditure using a
homogeneous panel of data for the US states. As a result, gross state product, the
proportion of the population over the age of 65 years, the degree of urbanization and
the number of hospital beds are the four key determining factors.
Erdil and Yetkiner (2009) have investigated the Granger-causality relationship
between real per-capita Gross Domestic Product and real per-capita health care
expenditure. Their findings show that the relevant type of Granger-causality is the
bidirectional one. The results show that one-way causality generally runs from income
to health in low and middle-income countries, whereas the reverse holds for high52
HEALTH CARE EXPENDITURE AND ECONOMIC GROWTH IN SOUTHERN ITALIAN REGIONS
income countries. Lin (2009) has studied the relationship between economic cycle and
health expenditures. By using data obtained from eight Asia-Pacific countries over the
period 1976 to 2003 and fixed-effects regression model, his results indicates that
unemployment rate is negatively and significantly correlated with total mortality and
mortality rates from cardiovascular diseases, motor vehicle accidents and infant
mortality. According to this empirical evidence, health might improve during economic
downturns. In addition, suicide is found to move counter-cyclically. The results also
show that unemployment affected mortality rates in an immediate and
contemporaneous way. Narayan (2009) has examined the behaviour of per-capita
health expenditures and per-capita GDP for 11 OECD countries, using a nonparametric test for two forms of asymmetries (deepness and steepness). The empirical
evidence underlines as, for six out of the 11 countries (the USA, the UK, Japan, Spain,
Finland and Iceland), either per-capita health expenditures or per-capita GDP are
characterized by asymmetric behaviour.
A lot of study – Kleiman (1974), Newhouse (1977), Leu (1986), Parkin et al.
(1987), Posnett and Hitiris (1992), Gerdtham (1992), Pritchett and Summers (1996),
Hansen and King (1996), Blomqvist and Carter (1997), Barros (1998), Roberts (1999),
and Narayan (2009) – have shown that a significant percentage of variation in per
capita health care expenditure (across countries and in time) could be explained by
variations in per capita GDP. This is often dubbed “direct causation”.
Yet, health expenditure also has an explanatory power on GDP, and this is dubbed
“reverse causation” (Rivera and Currais, 1999). Moreover, health determines school
participation and learning and hence human capital accumulation (Galor and MayerFaulkes, 2003). Another crucial assumption is that health care expenditures must have
positive effects on labour productivity (according to the “efficiency wages” hypothesis)
(Barlow, 1979; Srinivasan, 1992; Strauss, 1993; Behrman and Deolalikar, 1988;
Muysken et al., 2003; Schultz and Tansel, 1997; Glick and Sahn, 1998; Rivera and
Currais, 2005). However, misspecification problem occurs if causality is simultaneous
in both directions. Doing so, OLS estimation will produce biased and inconsistent
estimates of the structural parameters given that there is an endogenous relationship
between GDP and health care spending. Therefore, it is important to determine the
direction of the causality relationship between health care expenditures and GDP.
Furthermore, there is the possibility that the economy may respond asymmetrically
to positive shocks than to negative shocks (Beaudry and Koop, 1993).
Brenner conducted a series of studies (1971, 1975, 1979 and 1987) and found that
recessions and economic instability have a potentially adverse effect on health, while
subsequent studies wasn’t able to find an analogous empirical evidence (Wagstaff,
1985; Cook and Zarkin, 1986; McAvinchey, 1988; Joyce and Mocan, 1993).
Certain other studies supported the view that recessions are often accompanied by a
higher unemployment rate, increased psychosocial stress, declining income, reduced
psychological well-being: these effects lead to deterioration in both mental and physical
health. As a result, suicide was strongly associated with labour market conditions
(Yang and Lester, 1995; Viren, 1996; Lewis and Sloggett, 1998). Some recent works
underline as the total mortality rates were pro-cyclical, showing the trade-off between
53
COSIMO MAGAZZINO, MARCO MELE unemployment rates and mortality rates. The main the findings of these researches
provided evidence that health improves during economic downturns (Ruhm, 2000;
Laporte, 2004; Neumayer, 2004; Ruhm, 2004; Tapia Granados, 2005a, 2005b;
Gerdtham and Ruhm, 2006). On the contrary, Gerdtham and Johannesson (2005) found
that recessions increase the mortality rate for men, but don’t have any effect in relation
to women.
3. ECONOMETRIC METHODOLOGY AND DATA
Conventional regression techniques based on non-stationary time series produce
spurious regression and statistics may simply indicate only correlated trends rather than
a true relationship (Granger and Newbold, 1974). Spurious regression can be detected in
regression model by low Durbin-Watson statistics and relatively moderate R2. According
to Engle and Granger (1987), a linear combination of two or more non-stationary series
(with the same order of integration) may be stationary. If such a stationary linear
combination exists, the series are considered to be cointegrated and therefore long-run
equilibrium relationships exist. Incorporating these cointegrated properties, an ErrorCorrection Model (ECM) could be constructed to test for Granger causation of the series
in at least one direction. In this study, the ECM is specifically adopted to examine the
Granger causality between real GDP and electricity demand.
So, in order to investigate the stationarity properties of the series, the Augmented
Dickey-Fuller (ADF) (Dickey and Fuller, 1979, 1981), Phillips-Perron (PP, 1988),
Dickey-Fuller GLS (DF-GLS) (Elliott, Rothenberg and Stock, 1996), and Kwiatkowski,
Phillips, Schmidt, and Shin (KPSS, 1992) tests have been applied.
A time series that requires the first differencing filter to remove the stochastic trend is
called a time series that is integrated of order 1 (I(1)). Then we examine the unit root (or
stationarity) properties of the variables, accounting for structural breaks. The present
paper employs Zivot and Andrews (ZA, 1992) test to address this issue. Furthermore,
Clemente, Montañés and Reyes (CMR, 1998) developed a procedure allowing for a
gradual shift in the mean to test more than one break point.
The Johansen maximum likelihood procedure (Johansen, 1988; Johansen and
Juselius, 1990) is used for this purpose. Any long-run cointegrating relationship found
between the series will contribute an additional error-correction term to the ECM. The
short-run causality is based on a standard F-test statistics to test jointly the significance
of the coefficients of the explanatory variable in their first differences. The long-run
causality is based on a standard t-test. Negative and statistically significant values of the
coefficients of the error correction terms indicate the existence of long-run causality. For
the purpose of this paper, all the variables analyzed have been expressed in a logarithmic
scale. Our empirical study uses the time-series data of real GDP and real health care
expenditure for the 1980-2009 period in Southern Italian regions. Data are obtained from
54
HEALTH CARE EXPENDITURE AND ECONOMIC GROWTH IN SOUTHERN ITALIAN REGIONS
ISTAT1. The choice of the starting period was constrained by the availability of data on
health care expenditure data. In Table 1 variables of the model are summed up. All series
contain yearly data of the variables in real terms. As a preliminary analysis, some
descriptive statistics are presented in the following Table 2.
TABLE 1
List of the variables
Variable
GDP
HE
SOURCE :
Explanation
Real GDP, million EIT
Real health expenditure, million EIT
ISTAT data
TABLE 2
Exploratory data analysis
Variable
Mean
Median
Standard Deviation
GDP
79.059
80.435
15.480
HE
37.270
39.137
17.504
SOURCE : our calculations on ISTAT data
Skewness
-0.3901
-0.3574
Kurtosis
25.801
24.384
Range
70.049
75.266
The two series show a normal distribution, since Sk≅0 and K≅3.
4. DISCUSSION OF EMPIRICAL RESULTS
First of all, we obtained log-transformations of the time-series. The Inter-Quartile
Range shows the absence of outliers in our samples. Then, we applied time series
techniques on stationarity and unit root processes, in order to check some stationarity
properties. The correlation coefficients summarized in Table 3 indicate a high positive
correlation between real GDP and real health care expenditure, as well as between
GDP growth rate and health care expenditure variation.
TABLE 3
Correlation between public expenditure and price index
Region
Correlation coefficient between GDP and HE
Abruzzi
6,9125
Molise
6,9131944
Campania
6,9020833
Puglia
6,9152778
Basilicata
6,9152778
Calabria
6,9145833
Sicily
6,8986111
Sardinia
6,9118056
NOTES: Bonferroni adjustment applied
SOURCE: our calculations on ISTAT data
Correlation coefficient between ΔGDP and ΔHE
5,9923611
5,9125
5,9479167
5,9305556
5,8284722
5,8430556
5,9020833
5,9069444
Subsequently, Table 4 contains the results of stationarity and common unit root
1
See, for more details: http://www.istat.it/dati/db_siti/.
55
COSIMO MAGAZZINO, MARCO MELE tests for our variables. The second column presents results for Augmented Dickey and
Fuller (1979) test; the third one for Elliott, Rothenberg and Stock (1992) test; the
fourth column contains results for Phillips and Perron (1988) test; at last, in the fifth
column there are results for Kwiatkowski, Phillips, Schmidt and Shin (1992) test.
Here, results clearly indicate that real health care expenditure and real GDP are both a
I(1) process, in all eight Southern Italian regions.
TABLE 4
Results for stationarity tests
Stationarity tests
Deterministic
Region
Abruzzi
Molise
Campania
Puglia
Basilicata
Calabria
Sicily
Sardinia
Variable
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
component
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
intercept
ADF
NS: -2.602
LS: -3.583
DS: -3.414
DS: -3.165
NS: -2.471
LS: -2.705
DS: -3.455
DS: -3.265
NS: -2.518
LS: -3.279
DS: -3.430
DS: -4.117
NS: -2.408
LS: -3.568
DS: -3.436
DS: -3.919
NS: -2.235
LS: -2.868
DS: -3.567
DS: -3.579
NS: -2.538
LS: -3.860
DS: -3.414
DS: -4.105
NS: -2.527
LS: -3.434
DS: -3.439
DS: -4.038
NS: -2.450
ERS
NS: -0.357
NS: -0.028
DS: -2.687
DS: -2.298
NS: -0.390
NS: -0.137
DS: -2.686
DS: -2.557
NS: -0.330
NS: -0.096
DS: -2.765
DS: -3.404
NS: -0.373
NS: -0.060
DS: -2.756
DS: -3.442
NS: -0.374
NS: -0.147
DS: -2.846
DS: -2.282
NS: -0.333
NS: -0.508
DS: -2.672
DS: -3.163
NS: -0.341
NS: -0.105
DS: -2.817
DS: -3.454
NS: -0.334
PP
NS: -2.602
LS: -3.583
DS: -3.412
DS: -3.165
NS: -2.471
LS: -2.705
DS: -3.455
DS: -3.265
NS: -2.518
LS: -3.279
DS: -3.430
DS: -4.117
NS: -2.408
LS: -3.568
DS: -3.436
DS: -3.919
NS: -2.235
LS: -2.868
DS: -3.567
DS: -3.579
NS: -2.538
LS: -3.860
DS: -3.414
DS: -4.105
NS: -2.527
LS: -3.434
DS: -3.439
DS: -4.038
NS: -2.450
HE
intercept
LS: -3.433
NS: -0.096
LS: -3.433
NS: 1.22
ΔGDP
intercept
DS: -3.545
DS: -2.836
DS: -3.545
DS: 0.325
ΔHE
intercept
DS: -3.925
DS: -3.241
DS: -3.925
DS: 0.063
NOTES: LS: Level Stationary; NS: Non Stationary; DS: Difference Stationary.
SOURCE : our calculations on ISTAT data
56
KPSS
NS: 1.19
NS: 1.22
DS: 0.362
DS: 0.442
NS: 1.19
NS: 1.23
DS: 0.339
DS: 0.427
NS: 1.19
NS: 1.23
DS: 0.33
DS: 0.058
NS: 1.19
NS: 1.22
DS: 0.324
DS: 0.063
NS: 1.2
NS: 1.22
DS: 0.298
DS: 0.461
NS: 1.19
NS: 0.686
DS: 0.336
DS: 0.076
NS: 1.19
NS: 1.23
DS: 0.327
DS: 0.058
NS: 1.19
HEALTH CARE EXPENDITURE AND ECONOMIC GROWTH IN SOUTHERN ITALIAN REGIONS
TABLE 5
Results for additive outlier unit root tests
Region
Abruzzi
Molise
Campania
Puglia
Basilicata
Calabria
Sicily
Sardinia
Variable
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
HE
ΔGDP
ΔHE
GDP
SB
1993
1993
HE
ΔGDP
1993
1993
1993
1993
1993
1993
1993
1993
1994
1993
1994
1993
1993
1993
ΔHE
k
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
t-stat
-2.650
-2.706
-4.131
-3.944
-2.628
-2.464
-4.103
-3.941
-2.700
-2.591
-4.076
-4.129
-2.607
-2.723
-4.110
-3.884
-2.609
-2.532
-4.074
-4.374
-2.691
-2.713
-4.036
-4.010
-2.704
-2.633
-4.094
-4.078
-2.669
5% Critical
Value
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
-3.560
0
0
-2.623
-4.201
-3.560
-3.560
0
-3.842
-3.560
SOURCE : our calculations on ISTAT data
From the Table 5 above, we note that the Clemente et al. test results show that both
series are integrated of order 1 with a structural break. Yet, this additive outlier seems
to be occurred in 1993, probably because of new external commitment due to
Maastricht Treaty signature.
The lag-order selection has been chosen according to the final prediction error
(FPE), Akaike’s information criterion (AIC), Schwarz’s Bayesian information
criterion (SBIC), and the Hannan and Quinn information criterion (HQIC).
Cointegration tests have been subsequently applied, in order to be able to find the
57
COSIMO MAGAZZINO, MARCO MELE long-run relationship between GDP growth rate (ΔGDP) and health care expenditure
variation (ΔHE). As is shown in Table 6, Johansen and Juselius cointegration method
suggests that there is one cointegrating relationship in each region. In fact, the trace
statistic and the maximum-eigenvalue statistic reject r=0 in favour of r=1 at the 5%
critical value. As in the lag-length selection problem, choosing the number of
cointegrating equations that minimizes either the SBIC or the HQIC provides a
consistent estimator of the number of cointegrating equations. Yet, all these criteria
suggest a rank=1 for our data.
TABLE 6
Results for cointegration tests between GDP growth and health care expenditure variation
(ΔGDP and ΔHE)
Johansen and Juselius procedure
Region
Trace statistic
Maximumeigenvalue
statistic
Abruzzi
59.778
(9.42)
59.778
(9.24)
Molise
51.141
(12.25)
51.141
(12.52)
Campania
51.563
(9.42)
51.563
(9.24)
Puglia
62.711
(9.42)
62.711
(9.24)
Basilicata
89.631
(9.42)
89.631
(9.24)
Calabria
76.843
(12.25)
76.843
(12.52)
Sicily
54.239
(9.42)
54.239
(9.42)
Sardinia
51.927
(12.25)
51.927
(12.52)
SBIC
HQIC
AIC
-16.535
-19.568
-20.502
-14.950
-18.741
-19.909
-15.927
-18.960
-19.894
-16.576
-19.609
-20.543
-11.508
-14.541
-15.476
-15.608
-19.399
-20.568
-15.518
-18.551
-19.485
-13.483
-17.274
-18.443
Rank
r=1
r=1
r=1
r=1
r=1
r=1
r=1
r=1
NOTES: 5% Critical Values in parenthesis.
SOURCE : our calculations on ISTAT data
Granger causality tests suggest a bi-directional flow (with a feedback mechanism)
for health care expenditure and GDP in all Sothern Italian regions, in the long-run
(Table 7). On the other hand, in the short-run empirical findings roughly correspond to
58
HEALTH CARE EXPENDITURE AND ECONOMIC GROWTH IN SOUTHERN ITALIAN REGIONS
that of found for the long-run, but for two regions (Puglia and Basilicata) we find a
unidirectional causality, running from GDP to health care expenditure (as in Wagner’s
hypothesis).
TABLE 7
Results for short and long-run causality tests
Region
Lags
Log-likelihood
SBIC
Causality in the
long-run
Causality in the
short-run
Abruzzi
1
259.612
-12.361
GDP ↔ HE
GDP ↔ HE
Molise
3
349.668
-11.001
GDP ↔ HE
GDP ↔ HE
Campania
1
275.480
-13.803
GDP ↔ HE
GDP ↔ HE
Puglia
1
264.037
-12.763
GDP ↔ HE
GDP → HE
Basilicata
1
220.548
-0.8810
GDP ↔ HE
GDP → HE
Calabria
3
330.242
-0.9058
GDP ↔ HE
GDP ↔ HE
Sicily
1
271.442
-13.436
GDP ↔ HE
GDP ↔ HE
Sardinia
3
303.623
-0.9392
GDP ↔ HE
GDP ↔ HE
SOURCE : our calculations on ISTAT data
For all our equations, a Lagrange-multiplier (LM) test for autocorrelation in the
residuals of Vector Error-Correction Model (VECM) clarifies as at the 5%
significance level we cannot reject the null hypothesis that there is no serial correlation
in the residuals for the orders 1,…,5 tested. Checking the eigenvalue stability
condition in a VECM, the eigenvalues of the companion matrix lie inside the unit
circle, and the real roots are far from 1. As regard the Wald lag-exclusion statistics, we
strongly reject the hypothesis that the coefficients either on the first lag or on the
second lag of the endogenous variables are zero in all two equations jointly. The
Jarque and Bera normality test results present statistics for each equation and for all
equations jointly against the null hypothesis of normality. For our models, results
suggest normality. Finally, the analysis of ARCH effects shows the absence of this
problem for the estimated models.
5. CONCLUSIONS AND POLICY IMPLICATIONS
Using time series methodologies and studying the Southern Italian regions for 19802009 years, we obtain the result that real health care expenditure and real GDP
significantly affects each other in the long-run (as one cointegration relation exists). In
fact, based on a VEC model, we find that health care expenditure enters significantly
into the cointegration space. The short-run dynamics of the variables show that the
flow of causality runs from GDP to health care expenditure for Puglia and
Basilicata, whilst there is a short-run bi-directional causal relationship (or feedback
59
COSIMO MAGAZZINO, MARCO MELE effect) between the two series for the others (Abruzzi, Molise, Campania, Calabria,
Sicily and Sardinia).
Our empirical results are quite different to that of in Devlin and Hansen (2001),
while are in line with causality findings in Erdil and Yetkiner (2009).
Yet, if there is a bi-directional causal relationship, then economic growth may
demand more health care expenditure, whereas more health care expenditure might
also induce economic growth. So, health care expenditure and economic growth
complement each other. Consequently, we conclude that health care expenditure is a
limiting factor to GDP growth in Italy, and, therefore, shocks to the health expenditure
will have a negative effect on aggregate income.
6. SUGGESTIONS FOR FUTURE RESEARCHES
Further analysis may be conducted studying the nexus between different items of
health care expenditure and aggregate income in Italy.
COSIMO MAGAZZINO
MARCO MELE
ACKNOWLEDGEMENTS
Comments from Olga Marzovilla and Gian Cesare Romagnoli are gratefully
acknowledged. All remaining errors are entirely our responsibility.
This research has been financed by Futura onlus.
60
HEALTH CARE EXPENDITURE AND ECONOMIC GROWTH IN SOUTHERN ITALIAN REGIONS
REFERENCES
AA.VV., (2005), The contribution of health to the economy in the European Union, Health and
Consumer Protection.
BARLOW, R., (1979), Health and economic development: a theoretical and empirical review,
Human Capital and Development, 1, 45-75.
BARROS, P.P., (1998), The Black Box of Health Care Expenditure Growth Determinants,
Health Economics, 7, 533-544.
BAUMOL, W.J., (1993), Health care, education and the cost disease: a looming crisis for public
choice, Public Choice, 77 (1), 17-28.
BEAUDRY, P., KOOP, G., (1993), Do recessions permanently change output?, Journal of
Monetary Economics, 31, 149-163.
BEHRMAN, J., DEOLALIKAR, A., (1988), Health and Nutrition, in Chenery, H., Srinivasan, T.N.,
(eds.), Handbook of Development Economics, Elsevier Science.
BLOMQVIST, A.G., CARTER, R.A.L., (1997), Is health care really a luxury?, Journal of Health
Economics, 16(2), 207-229.
BRENNER, M.H., (1971), Economic Changes and Heart Disease Mortality, American Journal of
Public Health, 61(3), 606-611.
BRENNER, M.H., (1975), Trends in Alcohol Consumption and Associated Illnesses, The
American Journal of Public Health, 656(12), 1279-1292.
BRENNER, M.H., (1979), Mortality and the National Economy, The Lancet, 314(8142), 568573.
BRENNER, M.H., (1987), Economic Change, Alcohol Consumption and Heart Disease Mortality
in Nine Industrialized Countries, Social Science and Medicine, 25(2), 119-132.
BUKENYA, J., (2009), Do Fluctuations in Health Expenditure Affect Economic Growth?, The
Open Economics Journal, 2, 31-38.
CARRION-I-SILVESTRE, J.L., (2005), Health care expenditure and GDP: are they broken
stationary?, Journal of Health Economics, 24 (5), 839-854.
CEIS, (2008), Rapporto CEIS – Sanità 2008, Rome: CEIS.
CHAKROUN, M., (2009), Health care expenditure and GDP: An international panel smooth
transition approach, MPRA Paper, 17559, May.
COLOMBO, F., MORGAN, D., (2006), Evolution of Health Expenditure in OECD Countries,
Revue française des affaires sociales, April-September.
COOK, P.J., ZARKIN, G.A., (1986), Homicide and Economic Conditions: A Replication and
Critique of M. Harvey Brenner’s New Report to the U.S. Congress, Journal of Quantitative
Criminology, 2(1), 69-80.
CULLIS, J.G., WEST, P.A., (1979), The economics of health: An introduction, Oxford: Martin
Robertson.
DEVLIN, N., HANSEN, P., (2001), Health care spending and economic output: Granger causality,
Applied Economics Letters, 8(8), 561-564.
DI MATTEO, L., (2005), The macro determinants of health expenditure in the United States and
Canada: assessing the impact of income, age distribution and time, Health Policy, 71, 2342.
DI MATTEO, L., DI MATTEO, R., (1998), Evidence on the determinants of Canadian provincial
government health expenditures: 1965–91, Journal of Health Economics, 17, 211-228.
DREGER, C., REIMERS, H.E., (2005), Health care expenditures in OECD countries: a panel unit
root and cointegration analysis, International Journal of Applied Econometrics and
Quantitative Studies, 2, 2, 5-20.
61
COSIMO MAGAZZINO, MARCO MELE ERDIL, E., YETKINER, I.H., (2009), The Granger-causality between health care expenditure and
output: a panel data approach, Applied Economics, 41, 4, 511-518.
FELDSTEIN, P.J., (1988), Health care economics, New York: Wiley.
FORTE, F., MAGAZZINO, C., (2011), Optimal size government and economic growth in EU
countries, Journal of Analytical and Institutional Economics, XXVIII, 3, 295-321.
FREEMAN, D.G., (2003), Is health care a necessity or luxury? pooled estimates of income
elasticity from US state-level data, Applied Economics, 35, 495-502.
FUCHS, V.R., (1990), The health sector’s share of the gross national product, Science, 247, 534538.
GALOR, O., MAYER-FOULKES, D., (2003), Food for Thought: Basic Needs and Persistent
Educational Inequality, Centro de Investigacion y Docencia Economicas Division de
Economia Documento de Trabajo.
GERDTHAM, U.-G., (1992), Pooling international health expenditure data, Health Economics, 1,
217-231.
GERDTHAM, U.-G., JOHANNESSON, M., (2003), A Note on the Effect of Unemployment on
Mortality, Journal of Health Economics, 22(3), 505-518.
GERDTHAM, U.-G., JÖNSSON, B., (1991a), Price and quantity in international comparisons of
health care expenditure, Applied Economics, 23, 1519-1528.
GERDTHAM, U.-G., JÖNSSON, B., (1991b), Conversion factor instability in international
comparisons of health care expenditure, Journal of Health Economics, 10, 227-234.
GERDTHAM, U.-G., JÖNSSON, B., (2000), International comparisons of health expenditure:
theory, data and econometric analysis, 11-53, in Culyer, A.J., Newhouse, J.P., Handbook of
Health Economics, Elsevier Sciences.
GERDTHAM, U.-G., LÖTHGREN, M., (2000), On stationary and cointegration of international
health expenditure and GDP, Journal of Health Economics, 19(4), 461-475.
GERDTHAM, U.-G., RUHM, C.J., (2006), Deaths rise in good economic times: Evidence from the
OECD, Economics and Human Biology, 4(3), 298-316, December.
GIANNONI, M., HITIRIS, T., (2002), The regional impact of health care expenditure: the case of
Italy, Applied Economics, 34, 1829-1836.
GLICK, P., SAHN, E., (1998), Health and productivity in a heterogeneous urban labour market,
Applied Economics, 30, 203-216.
GRANGER, C.W.J., NEWBOLD, P., (1974), Spurious regression in econometrics, Journal of
Econometrics, 2(2), 111-120.
GROSSMAN, M., (1972), On the concept of health capital and the demand for health, Journal of
Political Economy, 80(2), 223-255.
HANSEN, P., KING, A., (1996), The determinants of health care expenditure: A cointegration
Approach, Journal of Health Economics, 15(1), 127-137.
HARTWIG, J., (2008), What drives health care expenditure? – Baumol’s model of ‘unbalanced
growth’ revisited, Journal of Health Economics, 27(3), 603-623.
HITIRIS, T., POSNETT, J., (1992), The determinants and effects of health expenditure in
developed countries, Journal of Health Economics, 11(2), 173-181.
JOHANSEN, S., (1988), Statistical analysis of cointegrating vector, Journal of Economic
Dynamics and Control, 12(2-3), 231-255.
JOHANSEN, S., JUSELIUS, K., (1990), Maximum likelihood estimation and inference on
cointegration with applications to the demand for money, Oxford Bulletin of Economics and
Statistics, 52(2), 169-210.
JOYCE, T.J., MOCAN, H.N., (1993), Unemployment and Infant Health: Time-Series Evidence
from the State of Tennessee, Journal of Human Resources, 28(1), 185-203.
62
HEALTH CARE EXPENDITURE AND ECONOMIC GROWTH IN SOUTHERN ITALIAN REGIONS
KLEIMAN, E., (1974), The determinants of national outlay on health, 66-81, in Perlman, M.
(ed.), The economics of health and medical care, London: Macmillan.
LAPORTE, A., (2004), Do Economic Cycles Have a Permanent Effect on Population Health?
Revisiting the Brenner Hypothesis, Health Economics, 13(8), 767-779.
LEU, R.R, (1986), The public-private mix and international health care cost in Culyer, A.J.,
Jönsson, B., (eds.), Public and private health services: Complementarities and conflicts,
Oxford: Basil Blackwell.
LEWIS, G., SLOGGETT, A., (1998), Suicide, deprivation, and unemployment: record linkage
study, British Medical Journal, 317, 1283-1286.
LIN, S.-J., (2009), Economic fluctuations and health outcome: a panel analysis of Asia-Pacific
countries, Applied Economics, 41, 4, 519-530.
MAGAZZINO, C., (2008), Modelli interpretativi della dinamica della spesa pubblica e “curva di
Armey”: il caso italiano, 1862-2001, Notizie di Politeia, XXIV, 92, 45-60.
MAGAZZINO, C., (2009a), Stima della spesa pubblica italiana secondo i modelli di Musgrave e
Rostow e di O’Connor, Il Risparmio Review, LVII, 3, July-September, 59-84.
MAGAZZINO, C., (2009b), Spesa pubblica disaggregata e “Legge di Wagner”, Economia,
Impresa e Mercati finanziari, 3, September-Decemeber, 7-31.
MAGAZZINO, C., (2009c), “Legge di Wagner” e spesa pubblica disaggregata: un approccio
VAR, Economia Pubblica, 5-6, 133-156.
MAGAZZINO, C., (2010a), “Wagner’s Law” in Italy: Empirical Evidence from 1960 to 2008,
Global & Local Economic Review, 14, 1, January-June, 91-116.
MAGAZZINO, C., (2010b), Dimensioni ottimali dell’operatore pubblico e crescita economica, Il
Risparmio Review, LVIII, 1, January-March, 5-33.
MAGAZZINO, C., (2010c), Wagner’s Law and Augmented Wagner’s Law in EU-27, MPRA
Paper, 26668, http://mpra.ub.uni-muenchen.de/26668/.
MAGAZZINO, C., (2011), GSP and Health Expenditure in Italian Regions, International Journal
of Business and Management, 6, 12, December, 28-35.
MCCOSKEY, S.K., SELDEN, T.M., (1998), Health care expenditures and GDP: panel data unit
root test results, Journal of Health Economics, 17 (3), 369-376.
MCAVINCHEY, I.D., (1988), A Comparison of Unemployment, Income and Mortality
Interaction for five European Countries, Applied Economics, 20(4), 453-471.
MEHRARA, M., MUSAI, M., AMIRI, H., (2010), The Relationship between Health Expenditure
and GDP in OECD Countries Using PSTR, European Journal of Economics, Finance and
Administrative Sciences, 24, 50-58.
MILNE, R., MOLANA, H., (1991), On the effect of income and relative price on demand for
health care, Applied Economics, 23, 1221-1226.
MUSHKIN, S.J., (1962), Health as an investment, Journal of Political Economy, 70(5), 129-157.
MUYSKEN, J., YETKINER, I.H., ZIESEMER, T., (2003), Health, Labor Productivity and Growth,
187-205, in Hageman, H., Seiter, S., (eds.), Growth Theory and Growth Policy, London:
Routledge.
NARAYAN, P.K., (2006), Examining structural breaks and growth rates in international health
expenditures, Journal of Health Economics, 25, 877-890.
NARAYAN, P.K., (2009), Are health expenditures and GDP characterized by asymmetric
behaviour? Evidence from 11 OECD countries, Applied Economics, 41, 4, 531-536.
NELSON, C.R., PLOSSER, C.I., (1982), Trends and random walks in macroeconomic time series:
Some evidence and implications, Journal of Monetary Economics, 10(2), 139-162.
NEUMAYER, E., (2004), Recessions Lower (Some) Mortality Rates, Social Science & Medicine,
58(6), 1037-1047.
63
COSIMO MAGAZZINO, MARCO MELE NEWHOUSE, J.P., (1977), Medical are expenditure: A cross-national survey, Journal of Human
Resources, 12, 115-125.
OKUNADE, A.A., KARAKUS, M.C., (2001), Unit root and co integration tests: time-series versus
panel estimates for international health expenditure models, Applied Economics, 33 (9),
1131-1137.
PARKIN, D., MCGUIRE, A., YULE, B., (1987), Aggregate health care expenditures and national
income: Is health care a luxury goods?, Journal of Health Economics, 6(2), 109-127.
PHILLIPS, P.C.B., (1986), Understanding spurious regression in econometrics, Journal of
Econometrics, 33(3), 311-340.
PIPERNO, A., DI ORIO, F., (1990), Social Differences in Health and Utilization of Health
Services in Italy, Social Science and Medicine, 31, 305-312.
POSNETT, J., HITIRIS, T., (1992), The determinants and effects of health expenditure in
developed countries, Journal of Health Economics, 11, 173-181.
POTRAFKE, N., (2010), The growth of public health expenditures in OECD countries: Do
government ideology and electoral motives matter?, Journal of Health Economics, 29, 797810.
PRITCHETT, L., SUMMERS, L.H., (1996), Wealthier is healthier, Journal of Human Resources,
31, 841-868.
RIVERA, B., CURRAIS, L., (1999), Economic growth and health: direct impact or reverse
causation?, Applied Economics Letters, 6, 11, 761-764.
RIVERA, B., CURRAIS, L., (2005), Individual Returns to Health in Brazil, 287-311, in LòpezCasanovas, G., Rivera, B., Currais, L., (eds.), Health and Economic Growth: Findings and
Policy Implications, Cambridge: MIT Press.
ROBERTS, A.A., (1999), The labour market consequences of family illness, Journal of Mental
Health Policy and Economics, 2, 183-195.
ROMAGNOLI, G.C. (2009), Diritto alla salute, sussidiarietà e assistenza sanitaria in Italia, Notizie
di Politeia, 96, 25-38.
RUHM, C.J., (2000), Are Recessions Good For Your Health?, Quarterly Journal of Economics,
115(2), 617-650.
RUHM, C.J., (2004), Maternal Employment and Adolescent Development, National Bureau of
Economic Research Working Paper, 10691.
SCHULTZ, T.P., TANSEL, A., (1997), Wage and labor supply effects of illness in Cote d’Ivoire
and Ghana: instrumental variable estimates for days disabled, Journal of Development
Economics, 53, 251-286.
SIMS, C., (1980), Macroeconomics and reality, Econometrica, 48, 1-48.
SPINKS, J., HOLLINGSWORTH, B., (2007), Cross-country Comparisons of Technical efficiency of
Health production: A Demonstration of Pitfalls, Applied Economics, November.
SRINIVASAN, T.N., (1992), Undernutrition: Concepts, Measurement and Policy Implications,
103-109, in Osmani, S., (ed.), Nutrition and Poverty, Oxford: Clarendon Press.
STRAUSS, J., (1993), The Impact of Improved Nutrition on Labor Productivity and HumanResource Development: An Economic Perspective, 149-170, in Pinstrup-Andersen, P.,
(ed.), The political economy of food and nutrition policies, Baltimore and London: Johns
Hopkins University Press for the International Food Policy Research Institute.
TANG, C.F., (2010), Multivariate Granger causality and the dynamic relationship between
health spending, income, and health price in Malaysia, MPRA Paper, 27298, December.
TAPIA GRANADOS, J., (2005a), Increasing Mortality During the Expansions of the U.S.
Economy, 1900-1996, International Journal of Epidemiology, 34(6), 1194-1202.
TAPIA GRANADOS, J., (2005b), Recessions and Mortality in Spain, 1980-1997, European
64
HEALTH CARE EXPENDITURE AND ECONOMIC GROWTH IN SOUTHERN ITALIAN REGIONS
Journal of Population, 21(4), 393-422.
VIREN, M., (1996), Suicide and business cycles: finnish evidence, Applied Economics Letters, 3,
737-738.
WAGNER, A.H., (1883), Finanzwissenschaft, Leipzig.
WAGSTAFF, A., (1985), Time Series Analysis of the Relationship Between Unemployment and
Mortality: A Survey of Econometric Critiques and Replications of Brenner’s Studies, Social
Science and Medicine, 21(9), 985-996.
WAGSTAFF, A., (1986), The demand for health: some new empirical evidence, Journal of
Health Economics, 5, 195-233.
WANG, Z. (2009), The determinants of health expenditures: evidence from US state-level data,
Applied Economics, 41(4), 429-435.
WANG, Z., RETTENMAIER, A.J., (2007), A note on cointegration of health expenditures and
income, Health Economics, 16(6), 559-578.
WHO, (2000), World Health Report 2010, World Health Organization.
YANG., B., LESTER, D., (2005), New directions for economics, Journal of Socio-Economics, 24,
433-446.
ZIVOT, A., ANDREW, D.W.K., (1992), Further evidence on the great crash the oil shock and the
unit root hypothesis, Journal of Business and Economic Statistics, 10, 251-270.
65
Pubblicazione fuori commercio
Registr. Tribunale di Napoli n. 178 del 15 luglio 1955
ISSN 0390-010X
Finito di stampare a Napoli nel mese di dicembre 2011
presso le Officine Grafiche Francesco Giannini & Figli S.p.A.
128